Sunday, February 20, 2005

Electronic Flight Bag

The electronic flight bag (EFB) is a concept for reducing or replacing an airline pilot’s flight bag with an electronic equivalent. ICM has implemented this concept in the form of an integrated product, for sale to commercial airlines, in cooperation with Intel, Jeppesen and IONA.

I have not looked at the detail of this product, so I cannot comment on it. What I want to explore in this post are some of the more general implications of this kind of product, in an SOA context.

From the product marketing material, I extract three points about the flight bag.
  1. It is a familar object in commercial aviation.
  2. Commercial airlines would benefit from automating this object. ("The aviation industry has long recognized the benefits of replacing paper in the cockpit with an electronic document delivery system.")
  3. However, this is a complex object that has previously resisted automation. (Reference to limitations in technology along with regulatory and standardization issues.)
In previous waves of automation we saw isolated paperwork systems being colonized by the computer, and then we saw links being built between these isolated systems. Complex and highly connected objects like the flight bag missed out on these waves of automation, for several interlinked reasons.

To automate the flight bag, we need to conceptualize it as the living centre of a highly complex system - with secure and reliable links into many other operational systems both inside the airline and within other organizations (airport? plane maintenance? air traffic control? GIS?). The electronic flight bag therefore becomes a centre within a network of centres. It is this kind of automation that this characteristic of the SOA wave.

And this kind of automation calls for a different kind of process. We may start with a familiar concept (the paperless xyz - in this case, the paperless cockpit). But we then need to model the implications of this paperlessness in terms of the multiple functions of paper in a given ecosystem. We then have to form a collaboration between several suppliers to compose a solution, that can then be composed in turn with the context of use in a given airline.

I have a mental image of something like a brain transplant - the transplanted brain will only work if you make all the right connections. Obviously there are technological advances (notably web services) that make this task a bit easier - but you have to think about the organizational implications as well, so it remains a challenge.

Although some of the marketing material describes the product as "complete", this appears to conflict with the statement that it is "designed for built-in integration and growth". From an SOA perspective, I hope the latter is true.

The electronic flight bag is a good step towards an SOA world (continuous network of services) if it satisfies four criteria:
  1. It creates a new service in its own right.
  2. It is composed of smaller services.
  3. It helps to complete one or more larger-scale services.
  4. It hints at future larger-scale services - it opens up future possibilities that we haven't yet fully formulated.

No comments: